Image Enhancement I
Image enhancement refers to tasks of sharpening image features, such as
edges, boundaries, or contrast, to make an image more suitable for visual
display and analysis.
The enhancement process increases the dynamic range of the chos
Color Image Processing
4 Color is a powerful descriptor that often simplifies object identification
and extraction from a scene.
4 In image analysis, the motivation for using color is that the human eye can
discern thousands of color shades and intensit
Image Processing and Analysis I
Prof. John Goutsias
1. For each of the image processors described below, determine whether the image processor is
(i) linear, and (ii) shift invariant
(a) g ( x , y ) = f ( x , y ) f ( x z , y )
Image Enhancement II
The histogram provides a global description of the appearance of an
image. It is a graph of the distribution of gray level values within an image.
The histogram p( rk ) , k = 0,1,., L 1, is a discrete function
Properties of the Fourier Transform of a
The Fourier transform is linear in the sense that
Fcfw_af (m,n)+bg (m,n)=aFcfw_f (m,n)+bFcfw_g (m,n)
If f ( m, n ) is a real-valued image, then its Fourier transform ex
Image Enhancement III
Enhancement by Spatial Filtering
Image enhancement can be achieved by means of a discrete image
processor T() , in which case the enhanced image g will be given by
g ( m, n ) = T( f )( m, n )
The processor may be linear or nonlin
THE HUMAN VISUAL SYSTEM
An objective of image processing and analysis is to develop techniques to
help a human observer interpret the content of an image. For this reason,
understanding human visual perception is important.
The human visual system is, b
This is the step subsequent to sampling in image digitization.
A quantizer maps a continuous variable u into a discrete variable ud ,
which takes values in a finite set
cfw_r1, r2 , rL
Define cfw_t k , k = 1,2,., L + 1
Image Processing and Analysis
Concerned with the manipulation and analysis of pictures by
Improvement of pictorial information for human interpretation.
Processing of scene data for autonomous machine perception.
The Continuous Image Processor
f (x,y )
g (x , y )
A two-dimensional continuous image processor is considered to be a
transformation T() of an input two-dimensional continuous image
f ( x , y ) to an output two-dimensional continuous image g( x, y).